Serving the Quantitative Finance Community

 
User avatar
BlueCell
Topic Author
Posts: 0
Joined: July 22nd, 2009, 3:16 pm

Extracting single name CDS spread

July 23rd, 2009, 10:30 am

Can someone please help me? I'm really desperate.How do I calculate the CDS spread if I have the following information:Hazard Rate: 2%Riskfree Interest Rate: 2%Recovery Rate: 40%Maturity: 5 yearsAnnual spread payments: 4Thank you,
 
User avatar
JoeyD123
Posts: 0
Joined: May 12th, 2009, 11:19 am

Extracting single name CDS spread

July 23rd, 2009, 5:08 pm

Either lay out or the cash flows and solve for the spread or... HazardRate = Spread/(1-RecoveryRate)
 
User avatar
BlueCell
Topic Author
Posts: 0
Joined: July 22nd, 2009, 3:16 pm

Extracting single name CDS spread

July 24th, 2009, 6:19 am

Thanks for your reply. Don't I have to include in the calculation the fact that the spread is being paid 4 times in a year?
 
User avatar
daveangel
Posts: 5
Joined: October 20th, 2003, 4:05 pm

Extracting single name CDS spread

July 24th, 2009, 6:48 am

Have u tried applying the Hull CDS model with constant interest rates and hazard rates ? I think it's tractable.
knowledge comes, wisdom lingers
 
User avatar
JoeyD123
Posts: 0
Joined: May 12th, 2009, 11:19 am

Extracting single name CDS spread

July 24th, 2009, 7:02 pm

QuoteOriginally posted by: BlueCellThanks for your reply. Don't I have to include in the calculation the fact that the spread is being paid 4 times in a year?You clearly have an academic problem as you have constant interest rates & a constant hazard rate. The quarterly pay vs. some other convention makes a small difference but would make more of a difference if the yield curve had significant slope and your hazard rate was non-constant (as in practice). For the fee leg, you multiply the fee (x/4) times the Disc Factor (e^(-0.02*t)) times the survival prob 1-e^(-0.02*t)) (here 0.02 is the hazard rate). The protection leg is 1- recovery times the default probability for the period (not cumulative default prob). You can use solver to find the spread which makes the fee and protection legs equal. My suggestion is to try to lay out the cash flows in this manner to help you gain some intuition. You should come out with a par spread close to the simple approximation I gave below.
 
User avatar
BlueCell
Topic Author
Posts: 0
Joined: July 22nd, 2009, 3:16 pm

Extracting single name CDS spread

July 26th, 2009, 8:39 am

Thanks Joey and Dave!I don't need an exact CDS spread, just an intuitive approximation:Using "Hazard Rate = Spread / (1 - Recovery)" with this data...Hazard rate: 1%Recovery: 20%Interest rate: 5%Annual payments: 4Time = 3 Years.. I get a CDS spread of 80 basispoints.However, if I run a computer program the spread is around 5 basispoints.That's a pretty big difference.
 
User avatar
daveangel
Posts: 5
Joined: October 20th, 2003, 4:05 pm

Extracting single name CDS spread

July 26th, 2009, 5:45 pm

I get a hazard rate of 2% with a flat spread curve of 1.10% per annum using your parameters. thats 2% rate, 40% recovery, 5 year CDS with quarterly payments
Last edited by daveangel on July 26th, 2009, 10:00 pm, edited 1 time in total.
knowledge comes, wisdom lingers
 
User avatar
ski
Posts: 1
Joined: October 25th, 2001, 8:26 pm

Extracting single name CDS spread

July 29th, 2009, 2:06 pm

you should also check out this site - will save you a lot of timehttp://www.cdsmodel.com/information/cds-model
 
User avatar
meforum
Posts: 0
Joined: August 19th, 2009, 2:43 am

Extracting single name CDS spread

October 2nd, 2009, 11:00 pm

QuoteOriginally posted by: JoeyD123Either lay out or the cash flows and solve for the spread or... HazardRate = Spread/(1-RecoveryRate)Apologies for what i am sure is a dumb question - new to CDS so bear with meIn all I read [OKane, JPM, etc] it states that a full bootstrap (using the CDS pricing function) is required to calculate the Survival Curve and Hazard RatesCould someone give me a little push on the HR=S/(1-R) intuition please. Sure I am missing something simple and the shoe will drop any time now......Thanks
 
User avatar
Ziggy
Posts: 1
Joined: January 27th, 2002, 10:59 pm

Extracting single name CDS spread

October 8th, 2009, 1:50 pm

HR=S/(1-R) assumes zero interest rates, and flat hazard curve throughout the period you are looking at. It´s a simple back-of-the-envelope method to get a quick hazard rate estimation. For example, with zero recovery, the formula simply says that the hazard rate is equal to the spread. The recovery is a simply scaling of the trade notional to the final payout upon default.When you relax the zero-interest assumption and start discounting, your fixed-leg (value of the payment upon default) is always going to decrease more than the coupon leg, as the default payment is a balloon at termination but coupon payments are annuities that start right away. As PV of both legs should match, this means that the higher interest rates you have, the higher the hazard rate to bring both legs in line. This difference will be really meaningful if we have a low spread name and high interest rates. It´s less important when we have vice-versa.Best regards,Z